Hydrogen-involved system safety detection method and system based on double-view background moire

Abstract

The present application relates to hydrogen safety visualization detection technical field, disclose a kind of hydrogen system safety detection method and system based on double view background schlieren.The method uses two optical field cameras, and collects the real-time background image of the flow field to be measured on hydrogen system under double view;Based on the real-time background image, obtain background spot displacement vector data;The background spot displacement vector data is converted, and two-dimensional plane gas concentration distribution data under each view is calculated based on optical fiber refractive index;Based on Radon transformation, two-dimensional plane gas concentration distribution data under double view is superimposed into three-dimensional gas concentration distribution field data;According to three-dimensional gas concentration distribution field data, hydrogen leakage detection is carried out to hydrogen system.The present application calibrates gas concentration by refractive index of light, with the advantages of low delay, high precision, accurate positioning leakage position.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydrogen safety visualization detection, in particular to a hydrogen system safety detection method and system based on a double-view background schlieren. BACKGROUND

[0002] Hydrogen has the characteristics of extremely flammable, explosive, colorless and odorless. Once the hydrogen pipeline in the hydrogen system leaks, the leakage position has a very high randomness, and the leakage concentration is not easy to detect quickly. The current commonly used gas detection means is basically a contact measurement method, and the current equipment measurement delay of the contact probe is generally 5-20s, and the detection of hydrogen concentration cannot guarantee high precision. In addition, the contact probe cannot quickly and accurately locate the leakage position, and once the hydrogen leaks and is ignited, it is easy to damage the probe. Therefore, the current research mainly focuses on a non-contact measurement method that can quickly detect hydrogen with high precision.

[0003] Background schlieren (BOS) technology is a large-view, non-contact quantitative flow field testing technology based on images. This technology combines particle tracking image processing technology (PIV), computer image processing technology and schlieren technology, compares distorted images with reference images, and obtains light deflection and density distribution of the observation area by calculating the offset of corresponding pixels. This technology not only continues the high sensitivity and high precision of schlieren technology for density field analysis, but also can observe a large field of view like PIV technology. Therefore, the technology is widely used in many research fields such as supersonic flow, combustion environment, plasma flow, etc.

[0004] Currently, for three-dimensional variable density flow, three-dimensional BOS measurement technology based on multi-projection tomographic reconstruction is commonly used. Tomographic BOS uses multiple cameras to obtain background displacement information at multiple angles, and through tomographic reconstruction algorithm, three-dimensional refractive index and density distribution can be obtained, but the cost is high and the system arrangement is complex, and the number of cameras, camera distribution and camera shooting angle will greatly affect the measurement effect. Image fiber BOS uses image fibers instead of cameras, and the image fibers that can take images at any projection angle are placed in front of the camera, so that a single camera can capture multiple instantaneous images at the same time. The system has low cost, but its spatial image resolution is poorer than that of traditional BOS and the system arrangement is complex. The emergence of light field BOS makes single-view volume imaging possible. A microlens array (MLA) is placed in front of a camera to replace multiple cameras. Compared with other methods, the light field BOS system is simple, but the image reconstructed by the single light field camera has the problem of lower axial resolution relative to lateral resolution, making it difficult to measure three-dimensional high-resolution flow fields. In addition, in the non-contact measurement of gas concentration, the three-dimensional tomographic technology is quite cumbersome for image post-processing, and the method of calibrating gas concentration by image grayscale has high limitations, making it difficult to apply in actual scenarios. SUMMARY

[0005] To solve the technical problems in the prior art, the application provides a hydrogen-related system safety detection method and system based on double-view background streaks.

[0006] To achieve the above object, the application provides the following technical scheme.

[0007] The application discloses a hydrogen-related system safety detection method based on double-view background streaks, comprising steps S1-S5.

[0008] S1. Two optical field cameras are used to collect real-time background images of a to-be-detected flow field above a hydrogen-related system under double views; wherein the two optical field cameras are arranged at the front view and the side view of the to-be-detected flow field, i.e. adjacent two sides, and are respectively perpendicular to two background plates arranged opposite to the to-be-detected flow field; the two background plates are both perpendicular to the horizontal plane, and are perpendicular to each other and have random dot arrays.

[0009] S2. Background spot displacement vector data is obtained based on the real-time background images.

[0010] S3. The background spot displacement vector data is converted, and two-dimensional plane gas concentration distribution data under each view is calculated based on the optical fiber refractive index.

[0011] S4. The two-dimensional plane gas concentration distribution data under double views is stacked into three-dimensional gas concentration distribution field data based on Radon transformation.

[0012] S5. The hydrogen-related system is detected for hydrogen leakage according to the three-dimensional gas concentration distribution field data.

[0013] As a further improvement of the above scheme, in step S2, the acquisition process of the background spot displacement vector data comprises:

[0014] The real-time background image of each view is subjected to Fourier transform with the corresponding pre-stored non-leakage background image, the vector displacement value between pixel points is identified by using the cross-correlation algorithm, and the background spot displacement vector data of the entire two-dimensional plane of each view is obtained by moving the window.

[0015] As a further improvement of the above scheme, step S3 comprises the following specific steps:

[0016] S31. The gas density ρ corresponding to each pixel in the real-time background image is calculated through the background spot displacement vector data. gas :

[0017]

[0018] In the formula, ρ air is the atmospheric density; K is the Gladstone-Dale coefficient; Δs is the displacement of any one background spot pixel, and s is the background spot pixel size;

[0019] S32. Calculate the light refraction index n air of the gas in the ambient gas: gas

[0020] n gas = 1 + Kρ gas

[0021] S33. Calculate the corresponding gas molar fraction X by the light refraction index n gas

[0022]

[0023] In the formula, n air is the atmospheric refraction index; T and T amb are temperature coefficients, T is 1.15, and T amb is 0.95;

[0024] S34. Obtain the gas concentration c corresponding to each pixel in the real-time background image according to the gas molar fraction X, so as to obtain the two-dimensional plane gas concentration distribution data under each viewing angle according to the gas concentration corresponding to all pixels in the real-time background image; wherein c = X·M, and M is the molar mass of hydrogen.

[0025] As a further improvement of the above scheme, in step S4, the two-dimensional plane gas concentration data under a specific single pixel under each viewing angle is decomposed into an n-term arithmetic sequence with 1 as the arithmetic progression based on Radon transformation, each term of the sequence is defined as a concentration eigenvalue, and the concentration eigenvalues are numbered to obtain n groups of concentration eigenvalue data on each pixel point in the viewing direction. By comparing the concentration eigenvalues at the same pixel point position in the two viewing directions, two concentration eigenvalues at each position in space are located, and the average of the two concentration eigenvalues is taken to obtain the gas reconstruction concentration at the position, so as to obtain the gas concentration values corresponding to each position of the to-be-measured flow field, and a three-dimensional gas concentration distribution field data is constructed.

[0026] As a further improvement of the above scheme, in step S5, the hydrogen leakage source is located according to the highest concentration position in the three-dimensional gas concentration distribution field data.

[0027] As a further improvement of the above scheme, in step S5, when it is determined that there is hydrogen leakage in the hydrogen-related system, the highest concentration position of the gas concentration field is marked, and an alarm signal containing the leakage source position information is sent to a remote interactive terminal. ​​

[0028] The application also discloses a hydrogen-related system safety detection system based on a double-view background schlieren method.

[0029] The image acquisition module is used for acquiring real-time background images of a to-be-detected flow field above a hydrogen-related system under double views; the image acquisition module comprises two light field cameras, which are arranged at a front view angle and a side view angle of the to-be-detected flow field, i.e. adjacent two sides, and respectively face two background plates arranged opposite to the to-be-detected flow field; the two background plates are both perpendicular to a horizontal plane, and are perpendicular to each other and have random dot arrays.

[0030] The data processing module is used for obtaining background spot displacement vector data based on the real-time background images; the background spot displacement vector data is converted to obtain two-dimensional plane gas concentration distribution data under each view angle; the two-dimensional plane gas concentration distribution data under double views is superimposed into three-dimensional gas concentration distribution field data based on Radon transformation; and hydrogen leakage detection is performed on the hydrogen-related system according to the three-dimensional gas concentration distribution field data.

[0031] As a further improvement of the above-mentioned scheme, the two background plates have fixed and mutually matched pixel positions, shape parameters and inverse scale parameters; and the distance between each light field camera and the corresponding background plate is adjustable.

[0032] Compared with the prior art, the application has the following beneficial effects:

[0033] Compared with the traditional sensor contact type measurement, the application has the advantages of high precision, low delay and wide test range, and the new double-view schlieren technology has higher axial resolution of light field imaging than the existing tomographic light field background schlieren three-dimensional flow field test method by superimposing the images of the front view angle and the side view angle. The hydrogen three-dimensional concentration field obtained by the double-view image acquisition method reduces the cost of the three-dimensional background schlieren system and simplifies the operation system.

[0034] In addition, the application proposes a calibration relationship between the background spot displacement and the gas concentration, which is more concise than the calibration relationship involved in the traditional three-dimensional flow field test, so that the system post-processing operation amount is significantly reduced. Therefore, the experimental device of the application is more simple, the gas concentration detection is more rapid, the leakage position of the hydrogen-related device after leakage can be accurately and rapidly positioned, and data support can be provided for hydrogen discharge action setting after leakage of the hydrogen-related system. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 The flowchart of the double-view background schlieren hydrogen-related system safety detection method in embodiment 1 of the application.

[0036] Figure 2 Schematic diagram of the arrangement of the light field camera and the background plate in embodiment 1 of the present application.

[0037] Figure 3 Principle diagram of the background streak technology. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0039] Embodiment 1

[0040] Referring to Figure 1 The present embodiment provides a hydrogen-related system safety detection method based on double-view background streaking, comprising steps S1-S5.

[0041] S1. Two light field cameras 1 are used to collect real-time background images of a flow field to be measured above a hydrogen-related system 3 under double-view angles; wherein the two light field cameras 1 are arranged at the front view angle and the side view angle of the flow field to be measured, i.e. adjacent to both sides, and are respectively perpendicular to two background plates 2 arranged opposite to the flow field to be measured; the two background plates 2 are both perpendicular to the horizontal plane, and are perpendicular to each other and have random dot arrays.

[0042] Referring to Figures 2 to 3 In step S1, the two background plates 2 have fixed and mutually matched pixel positions, shape parameters and inverse scale parameters; the distance between each light field camera 1 and the corresponding background plate 2 can be adjusted to be close or far. When collecting images, the high-speed camera and the background plate 2 are fixed to ensure that there is no slight shaking of the background during shooting, and the shooting angle of the camera and the background plate 2 are kept absolutely perpendicular.

[0043] S2. Obtaining background spot displacement vector data based on the real-time background images.

[0044] In step S2, the obtaining process of the background spot displacement vector data comprises:

[0045] Performing Fourier transform on the real-time background image of each view and the corresponding pre-stored non-leakage background image, performing cross-correlation analysis on the pre-stored non-leakage background image and the real-time background image using the particle high-speed velocity algorithm based on MATLAB, so as to identify the vector displacement values between the pixel points, and obtain the background spot displacement vector data of the entire two-dimensional plane of each view through the movement of the window.

[0046] In this embodiment, the collection method of the pre-stored non-leakage background image is as follows: when the hydrogen-related system 3 is not running, the camera is turned on to keep collecting images at all times, and the non-airflow background image collected at the same time of the main view angle and the side view angle is saved and stored in the system.

[0047] S3. The background spot displacement vector data is converted to obtain two-dimensional plane gas concentration distribution data under each view angle based on the refractive index of the optical fiber.

[0048] Step S3 includes the following specific steps:

[0049] S31. The gas density ρ corresponding to each pixel in the real-time background image is calculated by the background spot displacement vector data. gas :

[0050]

[0051] In the formula, ρ air is the atmospheric density; K is the Gladstone-Dale coefficient; Δs is the displacement of any one background spot pixel, and s is the size of the background spot pixel;

[0052] S32. The light refraction index n air of the gas with a gas density of ρ gas is calculated:

[0053] n gas = 1 + Kρ gas

[0054] S33. The corresponding gas molar fraction X is calculated by the light refraction index n gas :

[0055]

[0056] In the formula, n air is the atmospheric refraction index; T and T amb are temperature coefficients, T is 1.15, and T amb is 0.95;

[0057] S34. The gas concentration c corresponding to each pixel in the real-time background image is obtained according to the gas molar fraction X, so that the two-dimensional plane gas concentration distribution data under each view angle is obtained according to the gas concentration corresponding to all pixels in the real-time background image; wherein c = X·M, and M is the molar mass of hydrogen.

[0058] S4. The two-dimensional plane gas concentration distribution data under the double view angles is superimposed into three-dimensional gas concentration distribution field data based on Radon transformation.

[0059] In step S4, the two-dimensional planar gas concentration data of each specific single pixel under each view is decomposed into an n-term arithmetic sequence with 1 as the difference, each term of the sequence is defined as a concentration eigenvalue, and the concentration eigenvalues are numbered to obtain n groups of concentration eigenvalue data at each pixel point in the view direction. By comparing the concentration eigenvalues at the same pixel point position in the two view directions, the two concentration eigenvalues at each position in space are located, and the average of the two concentration eigenvalues is taken as the reconstructed gas concentration at the position, thereby obtaining the gas concentration values corresponding to each position of the flow field to be measured, and a three-dimensional gas concentration distribution field data is constructed.

[0060] S5. Hydrogen leakage detection of the hydrogen-related system 3 according to the three-dimensional gas concentration distribution field data.

[0061] In step S5, the highest concentration position in the three-dimensional gas concentration distribution field data is used to locate the hydrogen leakage source. When it is determined that there is hydrogen leakage in the hydrogen-related system, the highest concentration position of the gas concentration field is marked, and an alarm signal containing the leakage source position information is sent to the remote interactive terminal. The interactive terminal can be a mobile phone, a notebook computer, a tablet computer or other terminal devices that can interact with relevant personnel.

[0062] Embodiment 2

[0063] The embodiment provides a hydrogen-related system safety detection system based on double-view background schlieren, which applies the hydrogen-related system safety detection method based on double-view background schlieren in embodiment 1; the safety detection system comprises an image acquisition module and a data processing module.

[0064] The image acquisition module is used to acquire real-time background images of the flow field to be measured above the hydrogen-related system under double views; the image acquisition module comprises two light field cameras, and the two light field cameras are arranged at the front view and the side view of the flow field to be measured, i.e. adjacent two sides, and are respectively front to the two background plates arranged opposite to the flow field to be measured. The two background plates are both perpendicular to the horizontal plane, and are perpendicular to each other and have random dot arrays.

[0065] The two background plates have fixed and mutually matched pixel positions, shape parameters and inverse scale parameters; the distance between each light field camera and the corresponding background plate is adjustable.

[0066] The data processing module is used to acquire background spot displacement vector data based on the real-time background images; the background spot displacement vector data is converted to obtain two-dimensional planar gas concentration distribution data under each view; the two-dimensional planar gas concentration distribution data under double views is superimposed into three-dimensional gas concentration distribution field data based on Radon transform; and the hydrogen-related system is detected for hydrogen leakage according to the three-dimensional gas concentration distribution field data.

[0067] The above merely describes preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art, according to the technical solution and inventive concept of the present application, makes equivalent replacement or change within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A safety detection method for hydrogen-related systems based on dual-view background schlieren, characterized in that, include: S1. Using two light field cameras, real-time background images of the flow field to be tested above the hydrogen-related system are acquired from two perspectives. The two light field cameras are deployed on the frontal and side perspectives of the flow field to be tested, i.e., adjacent sides, respectively looking directly at two background plates arranged opposite to the flow field to be tested. Both background plates are perpendicular to the horizontal plane, perpendicular to each other, and have random dots. S2. Obtain background spot displacement vector data based on the real-time background image. The acquisition process includes: Perform Fourier transform on the real-time background image of each viewpoint and the corresponding pre-stored leak-free background image, use the cross-correlation algorithm to identify the vector displacement values ​​between pixels, and obtain the background spot displacement vector data of the entire two-dimensional plane for each viewpoint by moving the window. S3. The background spot displacement vector data is converted, and the two-dimensional planar gas concentration distribution data under each viewpoint is calculated based on the fiber refractive index. S4. Based on Radon transform, two-dimensional planar gas concentration distribution data from two perspectives are superimposed into three-dimensional gas concentration distribution field data; S5. Detect hydrogen leaks in hydrogen-related systems based on three-dimensional gas concentration distribution field data.

2. The method for safety detection of hydrogen-related systems based on dual-view background schlieren according to claim 1, characterized in that, Step S3 includes the following specific steps: S31. Calculate the gas density ρ corresponding to each pixel in the real-time background image using background speckle displacement vector data. gas : In the formula, ρ air Δs is the atmospheric density; K is the Gladstone-Dale coefficient; Δs is the displacement of any background spot pixel, and s is the size of the background pixel. S32. Calculate the gas density ρ air The refractive index n of the gas in the ambient gas gas : n gas =1+Kρ gas S33. Refractive index n of light gas Calculate the corresponding gas mole fraction X: In the formula, n air Atmospheric refractive index; T and T amb T is the temperature coefficient, with a value of 1.

15. amb The value is 0.95; S34. Obtain the gas concentration v corresponding to each pixel in the real-time background image based on the gas mole fraction X, and then obtain the two-dimensional planar gas concentration distribution data under each viewpoint based on the gas concentration corresponding to all pixels in the real-time background image; where c = X·M, and M is the molar mass of hydrogen.

3. The method for safety detection of hydrogen-related systems based on dual-view background schlieren according to claim 1, characterized in that, In step S4, the two-dimensional planar gas concentration data of a specific single pixel under each viewpoint is decomposed into an n-term arithmetic sequence with 1 as the arithmetic progression based on Radon transform. Each term of the sequence is defined as a concentration feature value, and the concentration feature values ​​are numbered to obtain n sets of concentration feature value data at each pixel point in the viewpoint direction. The concentration feature values ​​at the same pixel point position in two viewpoint directions are compared to locate the two concentration feature values ​​at each position in space. The average of the two concentration feature values ​​is taken to obtain the gas reconstructed concentration at that position, thereby obtaining the gas concentration values ​​corresponding to each position of the flow field to be measured, and constructing the three-dimensional gas concentration distribution field data.

4. The method for safety detection of hydrogen-related systems based on dual-view background schlieren according to claim 1, characterized in that, In step S5, the hydrogen leak source is located based on the location of the highest concentration in the three-dimensional gas concentration distribution field data.

5. The method for safety detection of hydrogen-related systems based on dual-view background schlieren according to claim 1, characterized in that, In step S5, when it is determined that there is a hydrogen leak in the hydrogen-related system, the location of the highest concentration in the gas concentration field is marked, and an alarm signal containing the location information of the leak source is sent to the remote interactive terminal accordingly.

6. A hydrogen-related system safety detection system based on dual-view background schlieren, characterized in that, The method for safety detection of hydrogen-related systems based on dual-view background schlieren as described in any one of claims 1 to 5 is applied; the safety detection system includes: An image acquisition module is used to acquire real-time background images of the flow field to be tested above the hydrogen-related system from two perspectives. The image acquisition module includes two light field cameras, which are deployed on the frontal and side perspectives of the flow field to be tested, i.e., adjacent sides, respectively looking directly at two background plates arranged opposite to the flow field to be tested. Both background plates are perpendicular to the horizontal plane, perpendicular to each other, and have random dots. The data processing module is used to acquire background spot displacement vector data based on the real-time background image; convert the background spot displacement vector data to obtain two-dimensional planar gas concentration distribution data under each viewpoint; superimpose the two-dimensional planar gas concentration distribution data under the two viewspoints into three-dimensional gas concentration distribution field data based on Radon transform; and perform hydrogen leakage detection on the hydrogen-related system based on the three-dimensional gas concentration distribution field data.

7. The hydrogen-related system safety detection system based on dual-view background schlieren according to claim 6, characterized in that, The two background panels have fixed and mutually matching pixel positions, shape parameters, and inverse scale parameters; the distance between each light field camera and the corresponding background panel is adjustable.

Images